Spaces:
Running
Running
File size: 8,153 Bytes
c4ac745 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 | import argparse
import json
import os
import shutil
import numpy as np
import pandas as pd
try:
from preprocessor import DataTransformer
from baselines.ClavaDDPM.preprocess_utils import topological_sort
except (ModuleNotFoundError, ImportError):
import importlib
import sys
base_dir = os.path.dirname(__file__)
full_path = os.path.abspath(os.path.join(base_dir, "..", "..", "preprocessor.py"))
spec = importlib.util.spec_from_file_location("preprocessor", full_path)
preprocessor = importlib.util.module_from_spec(spec)
sys.modules["preprocessor"] = preprocessor
spec.loader.exec_module(preprocessor)
DataTransformer = preprocessor.DataTransformer
def parse_args() -> argparse.Namespace:
parser = argparse.ArgumentParser()
subparsers = parser.add_subparsers(dest='op')
pre_parser = subparsers.add_parser('pre')
pre_parser.add_argument("--dataset-dir", "-d", default=os.path.join("data"))
# pre_parser.add_argument("--n-games", "-n", type=int, default=None)
pre_parser.add_argument("--out-dir", "-o", default=os.path.join("."))
post_parser = subparsers.add_parser('desimplify')
post_parser.add_argument("--dataset-dir", "-d", default=os.path.join("data"))
return parser.parse_args()
def main():
args = parse_args()
if args.op == "pre":
table_names = ["players", "courses", "course_maker", "plays", "clears", "likes", "records", "course_meta"]
players = pd.read_csv(os.path.join(args.dataset_dir, "players.csv"), sep="\t")
players = players.drop(columns=["image", "name"])
courses = pd.read_csv(os.path.join(args.dataset_dir, "courses.csv"), sep="\t")
courses = courses.drop(columns=["title", "thumbnail", "image"])
plays = pd.read_csv(os.path.join(args.dataset_dir, "plays.csv"), sep="\t")
clears = pd.read_csv(os.path.join(args.dataset_dir, "clears.csv"), sep="\t")
likes = pd.read_csv(os.path.join(args.dataset_dir, "likes.csv"), sep="\t")
records = pd.read_csv(os.path.join(args.dataset_dir, "records.csv"), sep="\t")
course_meta = pd.read_csv(os.path.join(args.dataset_dir, "course-meta.csv"), sep="\t")
all_plays = plays[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1)
clears = clears[
clears[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1).isin(all_plays)
].reset_index(drop=True)
likes = likes[
likes[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1).isin(all_plays)
].reset_index(drop=True)
records = records[
records[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1).isin(all_plays)
].reset_index(drop=True)
course_meta = course_meta[
course_meta["firstClear"].isna() |
course_meta[["id", "firstClear"]].astype(str).apply(
lambda row: "$$".join(row.tolist()), axis=1
).isin(clears[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1))
]
courses = courses[courses.maker.isna() | courses.maker.isin(players["id"])].reset_index(drop=True)
course_maker = courses[["id", "maker"]]
courses = courses.drop(columns=["maker"])
processors = {
table: DataTransformer() for table in table_names
}
if os.path.exists(os.path.join(args.out_dir, "processor.json")):
with open(os.path.join(args.out_dir, "processor.json"), "r") as f:
loaded = json.load(f)
for table in table_names:
processors[table] = DataTransformer.from_dict(loaded[table])
else:
processors["players"].fit(players, ["id"])
processors["courses"].fit(courses, ["id"])
processors["course_maker"].fit(course_maker, ["id"], ref_cols={
"maker": processors["players"].columns["id"],
"id": processors["courses"].columns["id"]
})
processors["plays"].fit(plays, ref_cols={
"id": processors["courses"].columns["id"],
"player": processors["players"].columns["id"],
})
processors["clears"].fit(clears, ref_cols={
"id": processors["courses"].columns["id"],
"player": processors["players"].columns["id"],
})
processors["likes"].fit(likes, ref_cols={
"id": processors["courses"].columns["id"],
"player": processors["players"].columns["id"],
})
processors["records"].fit(records, ref_cols={
"id": processors["courses"].columns["id"],
"player": processors["players"].columns["id"],
})
processors["course_meta"].fit(course_meta, ref_cols={
"id": processors["courses"].columns["id"],
"firstClear": processors["players"].columns["id"],
})
with open(os.path.join(args.out_dir, "processor.json"), "w") as f:
json.dump({
t: p.to_dict() for t, p in processors.items()
}, f, indent=2)
players = processors["players"].transform(players)
courses = processors["courses"].transform(courses)
course_maker = processors["course_maker"].transform(course_maker)
plays = processors["plays"].transform(plays)
clears = processors["clears"].transform(clears)
likes = processors["likes"].transform(likes)
records = processors["records"].transform(records)
course_meta = processors["course_meta"].transform(course_meta)
os.makedirs(args.out_dir, exist_ok=True)
os.makedirs(os.path.join(args.out_dir, "preprocessed"), exist_ok=True)
for table in table_names:
locals()[table].to_csv(os.path.join(args.out_dir, f"preprocessed/{table}.csv"), index=False)
os.makedirs(os.path.join(args.out_dir, "simplified"), exist_ok=True)
plays, course_meta, (clears, likes, records) = simplify_dataset(
plays, course_meta, clears, likes, records
)
for table in table_names:
locals()[table].to_csv(os.path.join(args.out_dir, f"simplified/{table}.csv"), index=False)
elif args.op == "desimplify":
desimplify_dataset(args.dataset_dir)
def simplify_dataset(plays, course_meta, *other_tables):
plays["playID"] = plays[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1)
course_meta["firstClear"] = course_meta.apply(
lambda row: np.nan if pd.isna(row["firstClear"]) else f"{row['id']}$${row['firstClear']}", axis=1
)
new_other_tables = []
for table in other_tables:
table["playID"] = table[["id", "player"]].apply(lambda row: "$$".join(row.tolist()), axis=1)
table = table.drop(columns=["id", "player"])
new_other_tables.append(table)
return plays, course_meta, new_other_tables
def desimplify_dataset(generated_dir):
plays = pd.read_csv(os.path.join(generated_dir, "plays.csv"))
course_meta = pd.read_csv(os.path.join(generated_dir, "course_meta.csv"))
def _process_df(df_name):
df = pd.read_csv(os.path.join(generated_dir, f"{df_name}.csv"))
df["index"] = df.index
merged_clears = df.merge(
plays[["id", "player", "playID"]], on="playID", how="left"
).set_index("index").loc[df.index].drop(columns=["playID"])
df = merged_clears
df.to_csv(os.path.join(generated_dir, f"{df_name}.csv"), index=False)
_process_df("clears")
_process_df("likes")
_process_df("records")
course_meta["index"] = course_meta.index
merged_course_meta = course_meta.merge(
plays[["id", "player", "playID"]].rename(columns={"id": "play_course_id"}),
right_on="playID", left_on="firstClear", how="left"
).set_index("index").loc[course_meta.index].drop(columns=["playID", "play_course_id", "player"])
merged_course_meta.to_csv(os.path.join(generated_dir, "course_meta.csv"), index=False)
if __name__ == "__main__":
main()
|